How do the neural circuits controlling movement combine sensory information, memories of past experiences, and internal state information (e.g. thirst, arousal, anger, fear) to produce even simple actions? And how does descending command activity change when a specific action (e.g. a lick in a specific direction) is made for different reasons (e.g. a reflexive vs. memory-guided lick)? Individual motor neurons receive thousands of inputs from across the brain and then integrate that information nonlinearly over dendritic arbors that span hundreds of microns. This complexity makes it difficult to assign a specific causative role to any single upstream circuit. Indeed, past studies have shown that many parallel circuits, from those involving cortex, the superior colliculus, the red nucleus, and simple reflexes in the brainstem, are each involved in driving similar orofacial movements. However, comparatively little is known about how these different pathways normally coordinate their activity with each other during normal behavior. One consequence of this gap in understanding pertains to disease. While many neurological impairments cause localized damage to movement-related brain regions (e.g. motor neuron degeneration in ALS, or the areas surrounding an infarction following a stroke), it is still unclear why some lesions cause permanent deficits while others can be compensated for by changes in the neural activity of other circuits in unaffected parts of the brain. The goal of this project is to discover the logic governing the coordination between different descending motor pathways, to determine how their recruitment depends on brain state (e.g. thirst or arousal), and to measure how such coordination changes following an acute or chronic brain injury. To do this, we will use recently developed large-scale neural recording technologies to interrogate many brain areas? including the motor nuclei themselves. First, using a new method for simultaneously monitoring many areas across dorsal cortex using Ca2+ imaging, I will explore the structure of neural activity contained in corticobulbar (i.e. cortex to medulla) projection neurons preceding both sensory and memory-guided directional lick bouts. Second, I will obtain electrophysiological recordings (using Neuropixels probes) from pyramidal tract neurons in motor cortex, the superior colliculus, and/or the medullar motor circuits themselves. Finally, I will study the acute and chronic effects of shutting down parts of the brain during licking behavior (using both optogenetic- silencing, and a mouse model of ALS to gradually kill motor neurons). These aims will be pursued at Stanford University, working under the co-mentorship of Karl Deisseroth and Surya Ganguli. This exceptional research community has an outstanding track record of both training postdoctoral fellows and successfully placing them in tenure-track faculty positions. My advisory committee will further ensure that I implement my training plan successfully and am able to establish my own lab to study how global brain computations drive specific actions.